An AI Readiness Assessment Should Leave You With More Questions, Not Fewer

We tend to think of assessments as things that give us answers.

You answer the questions, receive a score, identify some gaps and hopefully walk away knowing whether you're ready.

But I've started to think that a genuinely useful AI readiness assessment should do something else as well.

It should leave you with better questions.

Understanding creates questions

The more closely you examine an organisation, the more connections begin to appear.

You discover that a data issue is actually connected to a process.

A process problem turns out to be partly a technology problem.

A technology limitation is being compensated for by people.

A governance question leads to a question about permissions.

A conversation about permissions reveals that nobody is entirely sure where certain information lives.

One answer creates another question.

That isn't evidence that the assessment has failed.

It's evidence that you're beginning to understand the organisation properly.

A score can only tell you so much

There is value in measurement.

Knowing where strengths and weaknesses exist can help organisations prioritise, compare and make decisions.

But AI readiness isn't a single number.

Two organisations could receive exactly the same readiness score and have completely different circumstances behind it.

One might have excellent technology but fragmented information. Another might have strong data and processes but very little internal AI capability. Another might be technically sophisticated but lack clear governance or leadership direction.

The interesting part isn't simply where you scored.

It's why.

And then: what does that mean for us?

The questions become more specific

At the beginning of an AI readiness conversation, organisations tend to ask broad questions.

Which AI should we use?

Are we ready for Copilot?

Should we build an agent?

Can AI automate this?

What happens to our jobs?

But once an organisation understands its own environment, the questions become much more useful.

Which information should this AI be allowed to access?

Which process should we redesign before automating it?

Where would AI create the greatest value for us?

Which capability do our people need first?

What should we fix now, and what can wait?

What are we actually ready to do?

Those are harder questions.

But they're also much better questions.

Maybe clarity isn't having all the answers

I don't think AI readiness means reaching a point where an organisation suddenly has everything worked out.

AI itself is moving far too quickly for that.

Perhaps readiness is about having enough understanding to ask intelligent questions, recognise the implications of the answers and make deliberate decisions about what happens next.

A useful assessment shouldn't close the conversation.

It should make the next conversation considerably better.